Optimal load rating-based inspection planning of corroded steel girders using Markov decision process

Steel girder bridges are vulnerable to corrosion. To maintain their safety above a predefined target level, the load rating can be computed from the inspection results and guide the following maintenance actions. Optimizing inspection and maintenance based on load ratings has substantial practical and economic relevance. Load rating-based strategies can be categorized based on whether the inspection interval and replacement criteria are fixed or flexible. Existing studies focus on fixed inspection intervals throughout the service life. In general, their results are not optimal for inspection planning. To reduce life-cycle cost, aged steel girders may be inspected and repaired in an adaptive manner. To this end, a method based on Markov decision process (MDP) is proposed to compare the life-cycle cost of four load rating-based policies (i.e. uniform or adaptive non-uniform inspection interval, and fixed or adaptive replacement threshold). Load rating-based inspection planning is formulated as MDP and the optimal plans are obtained using dynamic programming. The conventional approach to discretize states cannot accurately approximate the non-stationary deterioration process, while state augmentation is successful in doing this but will increase computational cost. A comparison of two approaches is made to investigate their effects on life-cycle cost. A bridge girder under corrosion attack is used as an illustrative example. The results show that the load rating-based plan with an adaptive non-uniform inspection interval and fixed replacement threshold obtained using the state augmentation technique can be near-optimal.     

SAS Software Roadmaps

In this article we compute the expected Fisher information and the asymptotic variance–covariance matrix of the maximum likelihood estimates based on a progressively type II censored sample from a Weibull distribution by direct calculation as well as the missing-information principle. We then use these values to determine the optimal progressive censoring plans. Three optimality criteria are considered, and some selected optimal progressive censoring plans are presented according to these optimality criteria. We also discuss the construction of progressively censored reliability sampling plans for the Weibull distribution. Three illustrative examples are provided with discussion.     

Optimal Inspection Procedures

Five different optimum procedures are presented for inspection plans which do not directly measure the quality characteristic sought. Four of these plans are based on well-known criteria frequently used in applications of statistical decision theory. A fifth procedure maximizes, among other things, the probability that all defective items in a batch are removed among a fixed number, N, chosen for removal. This procedure replaces the problem of deciding which criterion to use with how many items one can afford to remove.     

基于结构可靠度的劣化桥梁维护方案成本优化研究

在桥梁管理系统中劣化桥梁结构的现有状态的评估和将来状态的预测是非常重要的内容.在预测过程中一方面应计算时变荷载和抗力,另一方面需考虑维持桥梁安全和服务水平的维护策略,这使得桥梁将来状态的预测变的十分复杂和困难.在劣化结构寿命评估和预测过程中要考虑各种参数的不确定性.在改进的结构劣化模型基础上,推导了在各种维护策略下的可靠指标评估公式,研究了维护方案与结构可靠指标的相互关系.提出了在分析寿命周期内的总成本中各单项成本的计算公式.根据寿命周期内成本现值最小的原则,在满足结构性能要求和预算限制约束下,决策出在分析周期内成本和性能都满足要求的维护方案.最后,以劣化的水泥混凝土桥面铺装为数值算例,证明了该方法的可靠性.     

Design of an integrated quality assurance strategy in production systems

High-quality of products is a critical issue for manufacturers to maintain their competitiveness in global markets. For this reason, more attention has been paid by operations managers and academics to the design of quality assurance strategies, acceptance sampling plans and inspection allocation problems. In the last decades, international research has studied and introduced several models and approaches to investigate these issues. The purpose of this paper is to provide a new methodology for designing and selecting correct integrated quality assurance strategies, defining cost models for acceptance policies and inspection station configurations. Generally, high-quality of items is guaranteed by avoiding defects, mainly caused by non-conforming components, resulting from instantaneous and standard infant mortality. Thus, an optimal acceptance policy is defined in order to reduce the instantaneous infant mortality defects. A closed-form equation has been introduced to determine easily and quickly the optimal percentage of checked items. Furthermore, a more convenient inspection station configuration is determined in order to minimise the expected total cost, composed of testing, inspection and penalty cost elements. The innovative concept of defect rate as an inspection time variable dependent has been introduced. The impact of different economic and survival parameters on designing inspection policies is also investigated. Finally, a real-life case study demonstrates the applicability of this methodology in real production systems and several considerations are reported about the future research, that the authors will carry out.     

烟用包装材料交验抽样方法研究

本文以国家抽样检验标准为依据，结合烟草行业自身特点，对制定烟用包装材料交验抽样方法的原则、依据和方法进行了较为深入的探讨和研究，以卷烟条与盒包装纸为例，设计了基于三个不同国家抽样标准下的三种交验抽样方案，通过对三种方案分别进行分析评价，给出了各自的使用条件和使用方法，对提升烟草行业烟用包装材料交验抽样工作的规范化和科学化水平具有一定的促进作用。     

Port-of-entry safety via the reliability optimization of container inspection strategy through an evolutionary approach

Up to now, of all the containers received in USA ports, roughly between 2% and 5% are scrutinized to determine if they could cause some type of danger or contain suspicious goods. Recently, concerns have been raised regarding the type of attack that could happen via container cargo leading to devastating economic, psychological and sociological effects. Overall, this paper is concerned with developing an inspection strategy that minimizes the total cost of inspection while maintaining a user-specified detection rate for “suspicious” containers. In this respect, a general model for describing an inspection strategy is proposed. The strategy is regarded as an (n+1)-echelon decision tree where at each of these echelons, a decision has to be taken, regarding which sensor to be used, if at all. Second, based on the general decision-tree model, this paper presents a minimum cost container inspection strategy that conforms to a pre-specified user detection rate under the assumption that different sensors with different reliability and cost characteristics can be used. To generate an optimal inspection strategy, an evolutionary optimization approach known as probabilistic solution discovery algorithm has been used.     

Multiobjective Decision Analysis for Acceptance Sampling Plans

In recent years there has been increased interest in the design and implementation of economically based sampling plans, especially those which consider a prior distribution of defectives in lots submitted for inspection. Sampling plans which are designed using this prior knowledge are often referred to as Bayesian sampling plans. This paper illustrates an approach to the design of Bayesian sampling plans which considers two criteria. The few previous attempts to design multiobjective Bayesian sampling plans have been based on expected values of the criteria. In this paper the authors present a design methodology which takes into consideration the distribution functions of the criteria after sampling inspection, and the utility function of the 'decision maker'. Illustrative examples of the use of the methodology are given as well as areas for further research.     

Optimal inspection of a finite population

This article presents a method for determining an optimal quality control (QC) inspection frequency for a manufacturing process where a specified number of items will be produced and where, if the process goes out of control and generates a defective item, it stays out of control. In addition, there is a QC inspection procedure than can detect a defective item. The frequency of inspection is based on minimizing the total cost. The total cost includes the cost of QC inspections plus the cost of manufacturing defective items. In this application the cost of manufacturing defective items is, after identifying a defective item, the cost of checking previously manufactured items until it is determined when the process went out of control, i.e., until a good item is found.     

Selecting a near‐optimal design for multiple criteria with improved robustness to different user priorities

In a decision‐making process, relying on only one objective can often lead to oversimplified decisions that ignore important considerations. Incorporating multiple, and likely competing, objectives is critical for balancing trade‐offs on different aspects of performance. When multiple objectives are considered, it is often hard to make a precise decision on how to weight the different objectives when combining their performance for ranking and selecting designs. We show that there are situations when selecting a design with near‐optimality for a broad range of weight combinations of the criteria is a better test selection strategy compared with choosing a design that is strictly optimal under very restricted conditions. We propose a new design selection strategy that identifies several top‐ranked solutions across broad weight combinations using layered Pareto fronts and then selects the final design that offers the best robustness to different user priorities. This method involves identifying multiple leading solutions based on the primary objectives and comparing the alternatives using secondary objectives to make the final decision. We focus on the selection of screening designs because they are widely used both in industrial research, development, and operational testing. The method is illustrated with an example of selecting a single design from a catalog of designs of a fixed size. However, the method can be adapted to more general designed experiment selection problems that involve searching through a large design space.     

Designing two mixed quick switching sampling plans for linear profiles

This study compares two proposed mixed quick switching sampling (QSS) plans for linear profiles as the quality characteristic. For the QSS plans, we recommend a binomial attribute plan for normal inspection and then a variable sampling plan for tightened inspection based on capability index C_puA of linear profiles with one-sided specifications. The difference between the two proposed QSS plans is in the tightened inspection. Tightened inspection of the first proposed plan is a single sampling using C_puA index, but tightened inspection of the second plan is a multiple dependent state repetitive (MDSR) plan based on C_puA index. The optimal parameters are obtained by nonlinear optimization. Simulation study for selecting parameters is conducted with various combinations of specified acceptable quality level (AQL), limited quality level (LQL), producer's risk, and consumer's risk. Simulation results confirm that the second proposed QSS plan which applies variable MDSR at tightened inspection performs better than another proposed plan. Hence, the approach of the second proposed plan is demonstrated in an illustrative example.     

Dependent Mixed and Mixed Repetitive Sampling Plans for Linear Profiles

In this paper, we present two new dependent mixed sampling plans and a mixed repetitive sampling plan based on the process yield index for linear profiles. A mixed sampling plan includes two stages; the first stage is a combined test by variables and by attributes, and the second stage is tested by attributes only. The first plan is based on the traditional mixed sampling scheme without marginal quality, and the second plan is based on the modified mixed sampling scheme with marginal quality. If a lot is very bad, the probability it can be rejected with the small initial sample is higher for dependent mixed sampling with marginal quality. We compare dependent mixed plans with other types of double sampling plans; the former outperforms the others with respect to average sample number. Additionally, the number of profiles required in the first stage is much smaller than in the second stage. When the inspection is costly and destructive, a mixed repetitive sampling plan includes one stage that combines the inspection by attributes and by variables and allows resampling in some conditions. We compare the mixed repetitive sampling plan with the repetitive sampling plans by attributes and by variables. The probability of acceptance using the mixed repetitive sampling plan performs better than the repetitive sampling plans by attributes and by variables. Three examples are used to illustrate the proposed methods. Copyright © 2017 John Wiley & Sons, Ltd.     

Inspection Error and Its Adverse Effects: A Model with Implications for Practitioners

Inspection error has clearly been shown to have adverse effects upon the results desired from a quality assurance sampling plan. These effects upon performance measures have been well documented from a statistical point of view. However, little work has been presented to convice the QC manager of the unfavorable cost consequences resulting from inspection error. This paper develops a very general, yet easily used, mathematical cost model. The basic format of the well-known Guthrie-Johns model is used. However, it is modified as required to assess the effects of attributes sampling errors of the first and second kind. The economic results, under different, yet realistic conditions, will no doubt be of interest to QC practitioners who face similar problems daily. Sampling inspection plans are optimized to minimize economic losses due to inspection error. Unfortunately, any error at all results in some economic loss which cannot be compensated for by sampling plan design; however, improvements over plans which neglect the presence of inspection error are possible. Implications for human performance betterment programs are apparent, as are trade-offs between sampling plan modification and inspection and training improvement economics.     

Lot Acceptance and Compliance Testing Using the Sample Mean and an Extremum

In industry, one sometimes compares a sample mean and minimum, or a mean and maximum, to reference values to determine whether a lot should be accepted. Particularly prominent examples of such procedures are “Category B” sampling plans for checking the net contents of packaged goods. Because the exact joint distribution of an extremum and the mean of a sample is usually complicated, establishing these reference values using statistical considerations typically involves crude approximations or simulation, even under the assumption of normality. The purpose of this article is to use the saddlepoint method to develop a fairly simple and very accurate approximation to the joint cumulative distribution function (cdf) of the mean and an extremum of a normal sample. This approximation can be used to establish statistically based acceptance criteria or to evaluate the performance of sampling plans based on criteria derived in other ways. These uses are illustrated with examples.     

Modeling Errors in Sampling Inspection: Effect of Degraded Performance

Studies have shown that inspection is not error-free and is systematically influenced by such factors as time available, the payoffs, complexity of the task, the decision goal, defect rate, and so on. This study shows how models of decision-making can be used to obtain better designs of sampling plans in the presence of human inspection error. This study also outlines a methodology to develop sampling plans under different levels of degraded human performance. Results revealed that sampling plans are sensitive to decision goals and the level of degradation.     

Analytical approximation for invasion and endemic thresholds,and the optimal control of epidemics in spatially explicit individual-based models

Computer simulations of individual-based models are frequently used to compare strategies for the control of epidemics spreading through spatially distributed populations. However, computer simulations can be slow to implement for newly emerging epidemics, delaying rapid exploration of different intervention scenarios, and do not immediately give general insights, for example, to identify the control strategy with a minimal socio-economic cost. Here, we resolve this problem by applying an analytical approximation to a general epidemiological, stochastic, spatially explicit SIR(S) model where the infection is dispersed according to a finite-ranged dispersal kernel. We derive analytical conditions for a pathogen to invade a spatially explicit host population and to become endemic. To derive general insights about the likely impact of optimal control strategies on invasion and persistence: first, we distinguish between ‘spatial'' and ‘non-spatial'' control measures, based on their impact on the dispersal kernel; second, we quantify the relative impact of control interventions on the epidemic; third, we consider the relative socio-economic cost of control interventions. Overall, our study shows a trade-off between the two types of control interventions and a vaccination strategy. We identify the optimal strategy to control invading and endemic diseases with minimal socio-economic cost across all possible parameter combinations. We also demonstrate the necessary characteristics of exit strategies from control interventions. The modelling framework presented here can be applied to a wide class of diseases in populations of humans, animals and plants.     

Optimal reliability, warranty and price for new products

The success of a new product depends on both engineering decisions (product reliability) and marketing decisions (price, warranty). A higher reliability results in a higher manufacturing cost and higher sale price. Consumers are willing to pay a higher price only if they can be assured about product reliability. Product warranty is one such tool to signal reliability with a longer warranty period indicating better reliability. Better warranty terms result in increased sales and also higher expected warranty servicing costs. Warranty costs are reduced by improvements in product reliability. Learning effects result in the unit manufacturing cost decreasing with total sales volume and this in turn impacts on the sale price. As such, reliability, price and warranty decisions need to be considered jointly. The paper develops a model to determine the optimal product reliability, price and warranty strategy that achieve the biggest total integrated profit for a general repairable product sold under a free replacement-repair warranty strategy in a market and looks at two scenarios for the pricing and warranty of the product. The model assumes that the sale rate increases as the warranty period increases and decreases as the price increases. The maximum principle method is used to obtain optimal solutions for dynamic price and warranty situations. Finally, numerical examples are given to illustrate the proposed model.     

Variables Sampling Plan for Resubmitted Lots in a Process with Linear Profiles

Acceptance sampling plans include a sampling scheme and a set of rules for determining whether an inspection lot from a supplier should be accepted or rejected. In some circumstances, the supplier is allowed to resubmit lots for further inspection when the original inspection result is unacceptable. In this study, two variables sampling plans based on the process‐yield index for a process with linear profiles are proposed to deal with lot sentencing. The single sampling plan is a special case of the resubmitted lots sampling plan. The plan parameters are determined using a nonlinear optimization method under the given values of producer's risk, consumer's risk, acceptable quality level, and lot tolerance percent defective. Numerous tables are provided to determine the plan parameters. One real example is used to illustrate our proposed method. Copyright © 2015 John Wiley & Sons, Ltd.     

Fault tolerance versus performance metrics for robot systems

The incorporation of fault tolerance techniques into robot systems improves the reliability, but also increases the hardware and computational requirements in the overall system. It is not always clear how to evaluate the merit, or ‘effectiveness’ of different fault tolerance approaches for a given application. In this paper, we present a new set of performance criteria designed to measure and compare the effectiveness of robot fault tolerance strategies. The measures, which are designed to evaluate fault tolerance/performance/cost tradeoffs, can also be used to evaluate pure performance or pure fault tolerance strategies. We show their usefulness using a variety of proposed fault tolerance approaches in the literature, focusing on multiprocessor control architectures.